1. IntroductionNaringin (NG) (4′,5,7—trihydroxy flavanone 7—rhamnoglucoside), a natural plant phenolic, is a flavanone-7-O-glycoside formed between the aglycon naringenin and the disaccharide neohesperidose, composed of a glucose and a rhamnose subunits (Figure 1). Its structure was explained for the first time in 1928 [1]. NG is mainly extracted from grapes and citrus fruits [2], the highest concentrations being found in the peel of grapefruit, Citrus paradisi (3.25%) and bitter orange, Citrus aurantium (2.11%) conferring them the bitter taste [1]. Studies revealed that the NG levels in plants decrease with their development towards maturity due to the conversion of this bioflavonoid into non-bitter compounds such as aglycone naringenin (NGN) [3]. NG was also detected in dried rosemary leaves (1261 mg/kg), in certain orchid-type leaves (110 mg/kg), peppermint (20 mg/kg), tomatoes (0.88 mg/kg) and other herb species [1].The increased interest for studying NG is based on the wide spectrum of its health benefits [2,4] due to its antioxidant, anti-inflammatory, bactericidal [5], anti-cancer, antimutagenic, cholesterol-lowering, and neuro- and cardiovascular-protective effects [6,7]. The presence of the –OH groups in the NG structure confers antioxidant properties to this bioflavonoid, but at higher concentrations, it exhibits pro-oxidant activities [8]. It was shown that NG reduces DNA damage by controlling the production of free radicals, with its radical-scavenging activity being dose-dependent [9]. The antitumoral activity of this phyto-compound in various cancers implies multiple mechanisms, some of them relying on its property to eliminate free radicals [10]. Due to its antioxidant and anti-inflammatory effects, NG alone [11] or in combination with trimetazidine [12] may exert protective activity against renal damage. Repeated administration of NG in mice induced anxiolytic-, antidepressant- and antiepileptic-like effects and increased locomotor activity, cognitive and memory performance via mechanisms including the enhancement of the antioxidant defense systems, the inhibition of lipid peroxidation, nitrosative stress and neuroinflammatory processes [13]. Studies revealed NG’s therapeutic effects in common musculoskeletal pathologies such as osteolytic and degenerative joint diseases, bones and joint infections [14,15]. Based on its ability to regulate the degree of reactive oxygen species, NG can help wound healing by favoring tissue regeneration [16]. Long-term consumption of NG presented no toxic effects on rats and humans, but after oral administration, it is poorly absorbed in the blood circulatory system due to its low bioavailability [15] which is caused by its reduced water solubility and permeability. Therefore, NG remains for a certain period in the gastrointestinal tract, where it is transformed into its main metabolite, NGN [17].Considering all the above-listed beneficial effects of NG on human health, which implies the importance of its daily intake from natural sources, mainly from citrus fruits, it is of interest to develop simple, fast and reliable methods for the detection of this bioflavanone and the quality control of its content or of the total antioxidants content in real samples. Various natural phenolics, some of them with very similar structures (e.g., flavanones and other flavonoids), coexist in different parts of the plants, and therefore, usually before the actual determination, the extraction [18,19] and/or separation [20,21] of NG from the sample matrix (e.g., plants, plant extracts, food, juices, etc.) was necessary. A recent review presents an overview of the extraction and sample preparation methods for NG chromatographic analysis from citrus fruits [1]. NG is a chiral compound and the ratio of the optically active isomers depends on the fruit maturation state [14]. Therefore, the chiral determination of NG from citrus peel and pulp was performed by ultra-performance liquid chromatography (LC) tandem mass spectrometry (MS) [22], while rapid resolution LC-MS coupled with isotope deuterium labeling was applied to the simultaneous quantification of NG and its metabolites NGN and 3-(4′-hydroxyphenyl) propanoic acid [17]. The determination of flavonoids, among them NG, from fruits and juices was also carried out by capillary electrophoresis with a diode array [20] or electrochemical detection using bare [23] or modified electrodes [24]. These techniques are selective and sensitive, but they are also expensive and time-consuming.Electroanalysis is a simpler, more rapid and cost-effective alternative for antioxidants and, consequently, NG determination, having the additional advantage that the instrumentation can be easily miniaturized and used for the assessment of small sample volumes. However, despite its electroactivity conferred by the –OH groups, NG is one of the flavonoids which was relatively little studied by electrochemical methods. A recent paper describes the use of an ion-exchange-based resistive sensor for the NG quantification between 1.72 × 10−6 and 1.72 × 10−4 mol/L and its removal from citrus juice [25]. An interdigitated electrode-based conductive sensor connected to a microcontroller platform, presenting a linear range of 25–100 ppm NG was applied to monitor the maturity stage of pomelo fruits by assessing their NG content [3].More selective electrochemical techniques are the voltammetric ones and the possibility of employing various modified electrodes enables the enhancement of both their sensitivity and selectivity. A thorough literature research pointed out that starting from 1998 till now there are relatively few studies regarding the electrochemical behavior and (volt)amperometric determination of NG [8,26,27,28,29,30,31], two of them being based on photoelectrochemistry [6,32]. Voltammetric techniques constitute also an important tool in the investigation of interactions between different bioactive molecules. NG square wave voltammetry with a carbon paste electrode was used to examine the NG interaction with DNA by monitoring the changes in the position and intensity of the adenine and guanine signals from oligonucleotides and dsDNA. The study revealed that at concentrations up to 1.00 × 10−5 mol/L (which is more than two times higher than the NG dose absorbable in an average human), NG acts as an antioxidant, having a genoprotective role, while above this level it could have pro-oxidant activity, favoring DNA oxidation [33]. One major drawback of the voltammetric methods is the fouling/passivation of the working electrode surface during measurements, e.g., by the formation of polymeric films, as is the case in the voltammetric analysis of polyphenols [34]. Consequently, in order to obtain reproducible results, the regeneration of the sensors’ electroactive surface area is compulsory. This involves an additional, tedious and time-consuming step, which can be eliminated by using disposable electrodes such as the pencil graphite electrode (PGE). In recent years PGE gained increasing applicability due to its similar or even better electrochemical characteristics in comparison to other commonly used working electrodes, besides its other inherent economic advantages (cost-effectiveness and easy availability) [35,36]. However, there are few reports using this electrode for the electroanalysis of bioflavonoids [34,37,38,39,40,41] which, based on their oxidation potential [42], are considered to have intermediate antioxidant power (AOP) [43]. To the best of our knowledge, there is no research that has been conducted on the NG electroanalysis of disposable, bare PGE. The results from this study for NG voltammetric quantification are comparable with some previously reported at other electrodes, with better results being obtained using chemically modified electrodes, as was expected. However, the favorable features of the PGE make it a versatile tool for the rapid and reliable screening of the total content of polyphenolics (TCP) with intermediate AOP of citrus fruits and their derivatives (e.g., juices). 2. Materials and Methods 2.1. Instrumentation

Voltammetric measurements have been performed on an Autolab PGSTAT 12 electrochemical system connected to a PC running the GPES4.9 software. A voltammetric cell containing a Ag/AgCl, KCl (3M) reference electrode, a Pt counter electrode and a pencil graphite electrode (PGE), if not stated otherwise, was the working electrode.

The tested working electrodes were different hardness, with PGEs of 0.50 mm graphite pencil leads (B, HB and 2B from the brands Laco and Rotring) having the geometrical active surface area (Ag) of 0.1589 cm2; a glassy carbon electrode (GCE) and a Pt electrode with the Ag of 0.0700 cm2 and 0.0314 cm2, respectively. Nevertheless, the main working electrode was the PGE, which consisted of 6.00 cm long, commercially available graphite pencil leads cut in half and introduced with the cut edge into a mechanical pencil lead used as a holder; 1.50 cm of the lead remained outside and always only 1.00 cm of this part was immersed in the analyzed solution so that the Ag of the electrode was always constant [41]. The electrical contact was realized by a copper wire soldered on the metallic part of the pencil. A new graphite pencil lead was employed for each measurement, excepting the studies requiring the recording of repetitive cyclic voltammograms.

In order to assure the reproducibility of the other solid electrode surface, the GCE and the Pt electrode were polished with 0.05 µm alumina powder on a special material, then rinsed with distilled water and air dried.

A pH/mV-meter Consort P901 Scientific Instrument (Belgium) equipped with a pH-sensitive combined glass electrode was used to measure the solutions’ pH.

Chromatographic analysis has been performed on a Shimadzu 20AD instrument with the following components: LC-20AD SP Shimadzu pumps with a maximum pressure of 400 bar, Shimadzu degasser DGU-20A5-Degasser, automatic injector, detector Shimadzu SPD-M20A Diode Array Detector, CTO-20AC Column Oven (thermostatic compartment of the column), LCsolution software. The separation was carried out on a chromatographic column of the Kromasil C18 type, 5 μm, 150 × 4.6 mm.

2.2. Reagents and Solutions

Naringin (NG) (≥95.0%, HPLC), naringenin (NGN) (>95%, p.a.), caffeic acid (≥98.0%, HPLC), gallic acid monohydrate (≥98.0%, ACS reagent), ethanol (≥99.5%, ACS reagent), potassium phthalate monobasic (KHPT) (ACS reagent), H3BO3 (1 g/tablet), CH3COOH (≥97%, ACS reagent), H3PO4 (≥85 wt. % in water, ACS reagent), Na2HPO4 × 2H2O and KH2PO4 (p.a., ACS reagent), NaOH (pellets), K3Fe(CN)6 (≥99.0%, ACS reagent) and KCl (99.0–100.5%, ACS reagent) were purchased from Sigma-Aldrich (Munich, Germany).

The 5.00 × 10−3 mol/L NG stock solution was daily prepared by dissolving the proper, accurate weighted amount of NG and diluting it with ethanol to the mark of a 10 mL volumetric flask. When not in use, this solution was refrigerated. The working solutions were obtained in 10 mL volumetric flasks by successive dilutions of the stock solution using an adequate supporting electrolyte. Acetate buffer solution (ABS; pH 3.80; 4.00 and 4.60), phosphate buffer solution (PBS; pH 7.00) and 0.05 mol/L (if not specified another concentration) KHPT (pH 4.00) were employed as supporting electrolytes. Britton–Robinson buffer (BRB) solutions with a pH between 1.81 and 11.20 were chosen for the investigation of the pH influence on NG voltammetric behavior. All aqueous solutions were prepared with deionized water.

2.3. Procedures

Cyclic voltammetry (CV) was applied to investigate NG voltammetric behavior, while differential pulse voltammetry (DPV) was used to establish the optimum conditions for its quantitative determination. The DPV peak currents were measured after the baseline correction was applied.

2.4. Real Samples Analysis

The applicability of the developed analytical tool on NG assessment in real samples was performed for pink and yellow grapefruit, being analyzed its peel and the fresh juice. Samples to be analyzed were obtained as follows: 40 (±0.026) g of vegetal material (fruit or peel) were weighed and further used for NG extraction. The peel was subjected to ethanolic extraction (100 mL) for 2 h, under continuous shaking, while fruits were squeezed, then a certain volume (20 mL) of juice was mixed with an equal volume of ethanol, shaken for 20 min for NG extraction. Prior to the voltammetric analysis, juice samples were filtered and diluted with supporting electrolyte (0.05 mol/L KHPT pH 4.00) to reduce matrix effects and bring the TCP within the linear range of the previously established voltammetric method for NG and NGN. HPLC-DAD-MS analysis of the alcoholic pectin precipitation from the fresh juice was carefully performed; the sample was filtered and centrifuged; the supernatant was used for analysis.

When determining NG content from juices, the standard addition method was used. 1.5 mL of NG standard solution of three different concentrations was added to the same volume of sample, and the DPV measurements were performed. The sample without any NG addition was also analyzed. The NG content was assessed using the values of the peak intensities. Three replicates were analyzed for each sample.